AGNC Investment Corp (AGNCN) 2011 Q4 法說會逐字稿

Good morning.

My name is Debbie and I will be your conference operator today.

At this time I would like to welcome everyone to the American Capital Agency shareholders call.

All lines have been placed on mute to prevent any background noise.

After the speakers' remarks there will be a question-and-answer session.

(Operator Instructions) Ms.

Katie Wisecarver, Investor Relations, please go ahead.

Thank you for joining American Capital Agency's fourth-quarter 2011 earnings call.

Before we begin I would like to review the Safe Harbor statement.

This conference call and corresponding slide presentation contains statements that to the extent they are not recitations of historical facts constitute forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995.

All such forward-looking statements are intended to be subject to the Safe Harbor protection provided by the Reform Act.

Actual outcomes and results could differ materially from those forecast due to the impact of many factors beyond the control of AGNC.

All forward-looking statements included in this presentation are made only as of the date of this presentation and are subject to change without notice.

Certain factors that could cause actual results to differ materially from those contained in the forward-looking statements are included in the risk factors section of AGNC's periodic reports filed with the Securities and Exchange Commission.

Copies are available on the SEC's website at SEC.gov.

We disclaim any obligation to update our forward-looking statements unless required by law.

An archive of this presentation will be available on our website and the telephone recording can be accessed through February 21 by dialing 855-859-2056 and the conference ID number is 42070592.

To view the Q4 slide presentation, turn to our website, AGNC.com, and click on the Q4 2011 Earnings Presentation link in the upper right corner.

Select the webcast option for both slides and audio or click on the link in the conference call section to view the streaming slide presentation during the call.

Participants on today's call include Malon Wilkus, Chairman and Chief Executive Officer; Sam Flax, Director, Executive Vice President, and Secretary; John Erickson, Chief Financial officer and Executive Vice President; Gary Kain, President and Chief Investment Officer; Chris Kuehl, Senior Vice President, Mortgage Investments; Peter Federico, Senior Vice President and Chief Risk Officer; Bernie Bell, Vice President and Head of Accounting and Reporting; and Jason Campbell, Senior Vice President and Head of the Asset and Liability Management.

With that I will turn the call over to Gary Kain.

Thanks, Katie, and thanks to all of you for your continued interest in AGNC.

The fourth quarter was characterized by a fair amount of uncertainty related to the European debt crisis and prepayment speeds.

As we will discuss on today's call, AGNC's portfolio continues to perform extremely well and we remain confident that the prepayment speeds on our assets should stay muted despite record low mortgage rates.

We also continue to have very little exposure to HARP 2.0, which is likely to begin to impact speeds over the next several months.

Now for the quarter, slow prepayments and improved valuations on lower loan balance and HARP securities helped to drive our very strong economic returns for both the quarter and for 2011 as a whole.

[We] remain steadfastly focused on actively adjusting our asset selection strategies as the market landscape and its mortgage valuations continue to evolve.

Before discussing our results in greater detail, I want to directly acknowledge the adjustment we announced yesterday to the first-quarter dividend.

We took this proactive action because we believed it is consistent with evolving market conditions and we are serious about transparency.

I will address the dividend in my prepared remarks as well as in the Q&A.

But I want to be clear that we remain extremely optimistic about our ability to produce very attractive returns for our shareholders.

Now with that let's turn to slides three and four where I can review a few of the highlights.

On page three we begin by presenting net comprehensive income.

We do this because GAAP net income per share has become less instructive given our decision at the end of last quarter to discontinue hedge accounting for our swaps.

As such, all of our hedges are now marked to market through GAAP income while price changes on our assets flow through the other comprehensive income line and shareholders' equity.

Net comprehensive income is essentially a complete mark-to-market earnings number, while GAAP income includes unrealized gains and losses on derivatives but not on our assets.

So our net comprehensive income was $2.27 per share.

This number is comprised of $0.99 of GAAP net income and unrealized gains on our assets, which are recorded in other comprehensive income of $1.28 per share.

Now net spread income, which many of you refer to as core income, was only $0.98 during the quarter or $1.01 when we back out approximately $0.03 per share of catch-up premium amortization.

This result was negatively impacted by a number of factors, including our mid-quarter equity raise, temporary increases in repo funding going into year-end, and high intra-quarter swap hedge ratios.

It is, therefore, not necessarily indicative of a future run rate.

Again, I want to remind investors that our asset yields and spreads are based on projections of lifetime prepayments rather than just actual speeds.

We are currently projecting an average CPR of 14% for our portfolio versus our actual quarterly CPR of 9%.

This projection is roughly 56% higher than what was realized in the quarter.

Now taxable net income, which is not impacted by changes in projected prepayment speeds or by unrealized losses on derivatives, remained strong at $1.61 per share.

Our undistributed taxable income increased during the quarter to $180 million, or $0.80 per share, after adjusting the share count given our Q4 equity raise.

As such, we again did not use any of our undistributed taxable income in paying our dividend.

Importantly, book value rose materially to $27.71 per share from $26.90 during the quarter.

Given the combination of the dividend and our strong book value gains, annualized economic or mark-to-market earnings were 33% for the quarter.

Now as you can see on the new slide, our mortgage portfolio increased to $55 billion by quarter end.

Leverage during the quarter was impacted by our equity raise and averaged only 7.6 times while year-end leverage came in at 7.9 times.

Importantly, our portfolio CPR remained very tame at 9% during the quarter despite significant prepayment increases on generic mortgages.

Moreover, our CPR, based on the numbers published in January, was only 8 CPR.

And hot off the presses in the prepayment report released by Freddie and Fannie just last night, our initial calculations show that our CPR was again only 8.

So now you have two of our three months of speeds for the first quarter and they are both 8.

Despite noticeably higher repo rates and a faster projected CPR, our net interest spread as of quarter end remained flat versus Q3 at 194 basis points.

This is a key point and shows that while we are in a lower spread environment, NIM at the end of Q4 would actually have been higher than as of September 30 had repo rates not temporarily spiked.

Lastly, we were very pleased with the timing of our October equity raise which was executed when our book value was at a local low point and mortgage prices had underperformed due to prepay and rate fears.

Since the equity raise we have seen significant book value appreciation.

Now let's turn to slide five and we can discuss the action that we took on the dividend.

Given that this is the first time in 2.5 years that we adjusted our dividend and given our practice of being as transparent as possible, we decided it was important to preannounce or announce this dividend change with our earnings release.

The Q1 dividend of $1.25 per share corresponds to a yield of approximately 17% at a stock price of around $29.50.

Clearly this is a very attractive return in today's market given a 10-year Treasury yield of around 1.9%.

Look, our proactive decision was based on our practice of establishing a dividend level that is consistent with market conditions, that allows us to distribute our taxable income in accordance with REIT rules, that is not expected to reduce our book value over time, and importantly that should not require regular quarterly adjustments over the near term.

Said differently, and this is a key point, we believe that the dividend should be sustainable for a reasonable period of time given current market conditions and our best estimate of future financial performance.

As we have done consistently in the past, we remain comfortable not immediately distributing all of our taxable income.

Maintaining some undistributed taxable income, which is currently at $0.80 per share, significantly reduces our need for quarterly dividend adjustments.

Now some of you may be thinking that we could have maintained the $1.40 for at least several more quarters and we don't disagree.

However, we felt that the new dividend is more consistent with current market conditions and should give investors more confidence in our ability to maintain the dividend.

Recent statements and actions by the Federal Reserve imply a funding and return environment characterized by somewhat lower spreads, but where attractive returns are likely to be available for a relatively long period of time.

Now lastly, I do want to stress that dividends are inherently uncertain and can be impacted by market conditions, changing financial performance, and other unforeseen developments.

And for a complete list of risk factors that can impact the business and therefore dividends, please refer to our recent 10-Qs and 10-Ks.

I anticipate that we will likely discuss this further in the Q&A portion of the call, so why don't we turn to slide six and take a quick look at AGNC's performance during 2011.

As we have stressed for the last few years, let's focus on the economic return analysis depicted on the top right.

In our opinion, this is the best measure of total value added for shareholders as it captures both dividends and changes in book value.

It is also a mark-to-market number where unrealized and realized gains are treated equally.

By this measure, AGNC produced a 37% economic return by growing book value over 14% and paying $5.60 per share in dividends.

This result exceeded these strong 32% economic return number Agency put up during 2010.

As you can see on the bottom right, a key driver of this performance was our continued ability to manage our exposure to prepayments despite falling mortgage rates and elevated speeds on generic mortgages during the second half of the year.

And while we are proud of these results, we recognize the opportunities and the risks that lay ahead may differ from those looking backwards.

And we are committed to actively managing our portfolio in an effort to appropriately balance the trade-offs between both risk and return.

But as Chris will discuss later, we believe our portfolio is very well-positioned as we enter 2012.

So with that said, let's turn to slide seven and look at what happened during the quarter.

The tables on the upper left show interest rate movements during the quarter were actually fairly muted for a change.

The swap curve flattened as the European debt crisis pushed shorter swap spreads wider, but shorter-term Treasuries had little room to rally further.

On the bottom left and on the top right you can see that fixed-rate mortgage prices were generally up between a few ticks and 0.5 percentage point.

Now some of you may be trying to reconcile our book value growth with this slide where mortgage prices really didn't move that much.

Well, if you turn to slide eight the picture should get a lot clearer.

As you can see in the graph, the prices of pools backed by HARP loans significantly outperformed more generic securities in the fourth quarter.

More specifically, the highest LTV HARP pools were 4.5% coupons, depicted in blue on both the graph and in the table, increased in price by a full percentage point more than generic mortgages.

Said another way, while 30-year 4.5% were one of the better performing coupons, going up almost 3/8 of a point in price, higher LTV HARP pools appreciated almost four times as much.

Lower loan balance pools, shown in gray, also performed very well during the quarter outperforming generic mortgages by around 0.5 point.

Now this incremental price appreciation on the types of assets that Chris highlighted for you on prior calls materially enhanced our book value performance during the quarter.

There is another thing really that this graph shows and that is that markets are not always efficient.

The payups, which are what the market calls the incremental price difference for HARP or loan balance pools versus TBA securities, were very low earlier in the year and remained relatively subdued, even when rates began to plummet during the third quarter.

Ironically, you could have bought HARP securities at a lower price than generic mortgages at the end of the first quarter of 2011.

So why is that?

Because back then the market was complacent around prepayment speeds and only cared about yield and the added liquidity provided by TBA or generic mortgages.

However, as rates fell and prepayment fears intensified the price performance of the slower prepay mortgages began to significantly outperform more generic stuff.

Furthermore, in the fourth quarter, when the beneficial speed performance was now undeniable and the HARP 2.0 changes were finalized, these price premiums took yet another leg up.

Now so far this quarter both generic mortgages and our specific holdings continued to perform very well.

Prices of generic 30-year 4.5%s are up more than 0.25 point in price since quarter end and some HARP and loan balance pools have appreciated almost double that amount.

While it's clearly early in the quarter, this performance has provided a tailwind to book value as we begin 2012.

So with that let's quickly look at slide nine which shows an updated version of the prepayment differences we have reviewed on a number of prior calls.

Consistent with the pricing chart we just went over, the prepayment on both HARP and loan balance securities continue to significantly outperformed those of more generic mortgages.

Mortgage rates, though, are back at their all-time lows and many leading indicators, such as the refi index, are back near their highs.

These factors, coupled with the changes in the HARP program, are likely to negatively impact speeds over the next several months, so this is certainly, again, no time to be complacent about prepayments.

With that, let me turn the call over to Chris to discuss how we have positioned the portfolio against this backdrop.

Thank you, Gary.

Let's turn to slide 10 to review the composition of our investment portfolio.

If you recall, during the third-quarter earnings call we discussed our extensive holdings and securities backed by lower loan balance loans and those originated through the HARP program.

During the fourth quarter, as Gary mentioned, these positions materially outperformed more generic TBA securities.

And while we still like pools backed by lower loan balances and loans originated through the HARP program, we selectively reduced our holdings given the significant improvement in valuations.

We increase our exposure to lower coupon 20-year and 30-year MBS as we felt it was both prudent to diversify and the relative risk and returns were compelling.

On average prepayment speeds continued to be very well contained despite record low mortgage rates.

Our projected average lifetime CPR increased from 13% CPR to 14% CPR due to a combination of lower rates and portfolio composition.

Let's turn to the next slide for more detail on why we continue to believe that we are well-positioned for the future.

Despite monetizing some of the outperformance in lower loan balance and HARP securities, our portfolio remains heavily weighted towards both of these strategies with more than 90% of our 15-year positioned backed by loans that were originated through the HARP program or have original loan balances less than or equal to $150,000.

Within the 30-year category nearly 70% of our holdings are backed by either HARP loans or loans with lower loan balances.

The Other category is comprised primarily of newly originated 30-year MBS with lower coupons to help mitigate prepayment risk.

And to further improve our disclosure this quarter we have added a more detailed breakdown of our projected lifetime CPRs.

While the projected speeds which drive our recorded asset yields are significantly higher than where the actual prepayment speeds are currently coming in, you have to remember that prepayment speeds do increase as loans season primarily due to housing turnover.

As we have discussed on prior calls, we continue to maintain our disciplined approach with respect to being willing to evolve the composition of our holdings in response to the changing market environment while maintaining responsible diversification.

With that I will turn the call over to our Chief Risk Officer, Peter Federico.

Thanks, Chris.

Today I will provide a brief update on our financing and hedging activities.

I will begin by reviewing our financing summary on slide 12.

As Gary mentioned, our repo cost increased significantly from 28 basis points in the third quarter to 40 basis points in the fourth quarter.

This 12 basis point increase was driven primarily by year-end funding pressures, which were somewhat stronger than usual due to concerns related to the European debt crisis.

Since year-end, balance sheet constraints have eased and repo rates have decreased about 10 basis points bringing our short-term borrowing costs back in line with the longer run average.

Additionally during the quarter, we extended the average maturity of our repo funding.

The weighted average original days to maturity of our repo funding increased significantly from 57 days at the end of the third quarter to 90 days at the end of the fourth quarter.

Looking ahead we will continue to source funding from across the full spectrum of the repo funding curve.

Turning to our hedging summary on slide 13, I will briefly discuss our derivative activity during the quarter.

First, our paid fixed swap portfolio increased slightly from $27 billion at the end of the third quarter to just over $30 billion at the end of the fourth quarter.

On average during the quarter our paid fixed swap portfolio covered 63% of our repo funding, a significant increase from the 52% in the third quarter.

As we discussed earlier, this increase in our swap portfolio was one of the factors that contributed to our lower net spread income in the quarter.

Despite the increase in our asset portfolio during the quarter, the notional balance of our put swaption portfolio remained unchanged at $3.2 billion.

The timing of our swaption activity is heavily dependent on the price of these options.

For most of the fourth quarter implied volatility, or said another way the market's expectations about the volatility of future interest rates, was very high.

This in turn caused the price of swaptions to be elevated for most of the quarter.

Late in the quarter, however, as implied volatility and swaption prices fell, we opportunistically replaced $1.2 billion of put swaptions that had expired earlier in the quarter.

Since year-end swaption prices have continued to fall and we, in response, have continued to grow our swaption portfolio.

As of the end of January our put swaption portfolio totaled $4.6 billion, a 44% increase from the year-end balance.

Before turning the call back over to Gary I would like to briefly discuss slide 21, which provides a reconciliation between GAAP net interest income and our non-GAAP measure, which we refer to as net spread income.

The key difference between these two measures relates to the geography of interest expense associated with our swap portfolio.

Prior to terminating hedge accounting at the end of the last quarter, the interest expense associated with these swaps was recorded in GAAP interest expense.

Today, broadly speaking, interest expense associated with these swaps that were previously in hedge relationships continues to be recorded through GAAP interest expense.

As shown on the table that expense was $53.6 million in the fourth quarter.

All other swap-related interest expense is recorded through other income.

That expense, as shown on the table, was $33.3 million in the fourth quarter.

The combination of these two numbers represents the total interest expense associated with our swap portfolio and, therefore, should be included in net spread income.

With that I will turn the call back over to Gary.

Thanks, Peter.

As you guys can see in the left-most columns of slide 15, our net spreads as of quarter end remained unchanged from September 30 at 194 basis points.

This result includes the roughly 12 basis point higher repo costs embedded in the year-end result that Peter mentioned.

Our 194 basis point NIM is also based on our 14% prepayment projection and would have been closer to 210 basis points had we used actual prepayment speeds.

The 210 basis point number is likely more comparable to our peers, which we believe use actual quarterly CPRs.

Also of note is the decline in our total operating expenses, which are now down to less than 160 basis points.

Now as you can see at the bottom of the page, our net ROE, as of December 31, totaled just under 17%.

Now remember, this number does not include any assumption around realized gains and includes the elevated repo costs.

So now let's turn to page 16 so I can conclude by giving you a little more transparency into what we think the future might hold as we begin 2012.

First, as I mentioned earlier, we remain optimistic that we can continue to produce strong results for our shareholders.

We continue to believe that the prepayments on our portfolio will remain well-behaved and we have given you two months of speeds for this quarter so far.

We also have very little HARP 2.0 exposure and our asset yields already incorporate significant increases in speeds versus where our portfolio is currently prepaying.

In addition, as Peter mentioned earlier, repo rates have come down close to 10 basis points and the Fed has communicated its expectation that short rates will remain near zero until at least the latter part of 2014.

One other thing to keep in mind is the benefit our book value increases have had on our earnings power.

As book value grows it facilitates a larger portfolio and greater earnings, even assuming that leverage remains the same.

As such the $5.48 of incremental book value gains that we have achieved over the past two years makes generating a fixed amount of earnings per share considerably easier than it would have been in the past.

So with that let me turn the call back to the operator and open the lines for questions.

(Operator Instructions) Arren Cyganovich.

Thank you.

Just kind of drilling down onto the spread income, I know you kind of walked through some of the benefits that you will have in terms of lower cost of funds with repo getting back to normal.

But can you talk a little bit about the reconciliation of why the $1.01 spread income is not a good run rate for the quarter relative to the 125, and what pieces of your portfolio will actually help you reach that more sustainable dividend level without, I guess, the benefit of realized taxable gains that you have in the quarter?

Sure.

Look, I think what I would tell you -- first off, as we said on the call, we did do a mid-quarter equity raise.

And one thing that was a little different about this equity raise is clearly the market environment in Q4 was more volatile than, let's say, it has been in the past.

So in the past we intend to pre-buy securities for our equity raises and so you didn't see as much of an impact due to the time lag for purchases.

Clearly, given the market environment in the fourth quarter we would have been less comfortable and were less comfortable doing that.

So that is one factor.

Clearly, you had changing composition of the swap book relative to the asset book and timing differences there.

We obviously talked about repo rates.

But I think more importantly for you in terms of getting a feel for kind of we will call it our economics would be -- let's look back to page 15 for a second, because it has our asset yield as of 12/31, which is 3.07%.

It has our cost of funds.

Again, this is the elevated cost of funds of 113, which brings you down to a net margin of 194 basis points.

When you apply the leverage that we had then that brings you up to a gross ROE of 18.5%.

Again, without any benefits of active management or UTI or any of those issues, you are looking at a net ROE that is pretty close to 17%.

If you then just look at the repo rates and say -- that is maybe not for the first quarter but when they normalize given the decline we have seen, if they normalize by 10 basis points that will bring that net ROE probably to the mid-17%s.

And if you just do a mathematical calculation on a mid-17%s ROE with book value of $27.71 that gets you to the low $1.20s in earnings per share.

So that is not a forecast.

It's just a mathematical calculation of defining the net ROE, but I think that is a much better starting point than looking at a noisy intra-quarter number.

That is helpful.

Then also what are you seeing in terms of spreads on new investments that you are putting on relative to the 194 that you ended up at the end of the quarter?

Surprisingly, they are not really that different.

I think if you can find attractive mortgage securities you are still in the area of a couple hundred basis points so to speak.

And the reason for that is 30-year mortgages are in the low 200s, maybe 225.

Again, if you can pick the securities.

15 years are clearly lower than that and more like in the 1.5 to 1.75 area depending upon which coupons.

So the returns are fine on new purchases given the fact that swap rates are very low.

I think what you just have to be cognizant of is that we continue to be in a lower rate environment where you have to continue to be very selective around what assets you add.

Thank you.

Jason Weaver, Sterne, Agee.

-- on getting that capital to work so quickly.

Gary, just on the tail of your comment regarding not using the undistributed income to maintain the dividend, I am just trying to understand.

If the purpose as you backed up is to avoid making quarterly changes to the payout going forward, how does that not apply now?

And am I to understand that the 125 level is your assumption of a long-term sustainable rate?

So let me take that in two pieces.

I think the first piece of your question was you cut the dividend even though you had lots of UTI, so what does UTI mean?

Look, what I would say is UTI, or undistributed taxable income, allows us to continue to pay dividends even if taxable income disappoints for a few quarters.

Therefore, it would support us paying a dividend that was otherwise consistent with market conditions and not really have to worry about, again, a weaker quarter here or there.

So it really removes taxable income as a constraint for some period of time on the dividend, but it doesn't mean -- and if you look at our dividend paying history, it's never meant that we are going to maintain a particular dividend until we have spent all of our undistributed taxable income.

That is not the way we look at it.

And the key drivers of the new dividend, again which is roughly 17%, is around being consistent with the current environment as we see it and having a dividend that I think investors can have more comfort in relative to the $1.40.

Our hope is the $1.25 ends up being very similar to what $1.40 was two-and-a-half years ago.

Got it, thank you.

And just turning to slide 11 where you detail the differing measures of prepayment protection.

I notice that you don't list any of the 20-year securities, but in the reference it looks like most of those are either -- well, they are both either low coupon and relatively new origination type stuff.

So I am just wondering why that wasn't listened as newly seasoned low coupon as limited prepayment risk?

Well, we chose two strategies to highlight in those -- on that page, and so we really took the bigger portions of the portfolio -- 20-year is roughly 10% -- and so we detailed really along the two main strategies that we have talked about historically.

To your point, we have added 20-year mortgages and those are kind of very new low coupons but they are not low loan balance for -- there is very little low loan balance or HARP securities in there.

There are some.

But I want to be clear, even in some of the other categories, in 15-year and 30-year, there are some securities that offer kind of favorable prepayment attributes.

So as an example, there are some securities with loan balances that are between 150 and 175 that we actually include in the other category.

There are other things that we look at, such as the FICO score of borrowers and other kind of LTV-related issues or geographics that can go in there.

So you shouldn't assume that anything that is not in loan balance or in HARP is just a generic mortgage security that we are nervous about.

So hopefully that helps.

Yes, yes, definitely.

Then finally one last question.

Looking in your duration gap table, and I know the limitations of modeling something like that, but with the zero net duration it seems that it's printing -- is your expectation that you are expecting rising long-term rates or is the reasoning that the duration estimates look artificially low on your prepay protected pools?

So with respect to we were expecting higher rates, the answer is no.

It's not that we are -- we really want to and we have stressed this in the past, and I think Peter has highlighted this over the last few calls since he has joined us as well, we are not trying to make big picture calls on interest rates.

We need to have and we are dedicated to having our portfolio perform in either case.

What is interesting is that when interest rates are at lows like today and when mortgage durations or the price sensitivity of mortgages are at their lows, like near two years, then you don't want to run a big duration gap from a risk perspective.

Why?

Because the risk is much more so to the durations getting longer.

So if interest rates went up 200 basis points or more, the duration of this portfolio may go to near five years.

And so if you think of mortgages as having this range of 2 to 5 or 1.5 to 5 or whatever you want to apply, the closer you get to the low end the less willing you should be from just an overall risk perspective to run a large duration gap.

And I think that is really the key driver of the positioning rather than our view on interest rates.

Got it.

Okay.

Well, thanks a lot, guys.

Bill Carcache, Nomura.

Good morning.

Thank you for taking my question.

Can you tell us what portion of the $136.6 million loss on derivative instruments in the P&L is unrealized?

Yes, this is Peter.

Let me give you a quick breakdown of that.

There is really three components to that $136 million number.

One is the $33 million that we referred to on the table on page 21 related to our interest expense associated with our swaps.

There is about another $17 million associated with realized gains/losses; it's a realized loss associated with TBA and treasury hedges.

And then the residual, which is about $86 million, is the unrealized loss associated with our swap portfolio.

So I appreciate that you are no longer applying hedge accounting, but from the standpoint of trying to strip out the impact of unrealized gains and losses, if we are trying to do that what are your thoughts on essentially taking the $86 million unrealized out of the reported number to get to something closer to, I guess, a little bit more of an apples-and-apples number since that won't include the effect of unrealized gains and losses on either the swaps or the asset side of your balance sheet?

Well, I think that is one way of approaching it.

I think, though, that if you start maybe looking at page 21, slide 21.

I think it might be easier, if you are trying to get at the all-in cost from an interest carry perspective associated with our swaps.

I would look at two numbers on slide 21, which is the $53.6 million and the $33 million.

The combination of those two numbers is the full interest expense on our swap portfolio regardless of whether the swap was ever in a hedge relationship or not.

So together that $87-or-so-million would be the full cost, the interest cost, of the swap, the pay leg versus the received leg on the $30 billion of swaps that we have.

Got it.

Just finally a big picture question.

Do you think that the proceeds from a potential settlement between the banks and the AGs could be used to refinance borrowers from the nonagency mortgage market into the FHA market?

And more broadly, what do you think that could mean for the market and the agency mortgage REITs in particular?

Well, the reality is it's not really going to affect the agency market very much in that anything on that front would relate to prepayments on nonagencies, which generally speaking for nonagencies is a good thing.

The only kind of ramification would be, if FHA or Ginnie production picks up, could that impact the technical factors for the agency market.

And the short answer is very little.

There is basically negligible new supply of agency mortgages and I think that supply would be easily absorbed and it wouldn't be that big picture of an issue.

Generally speaking, just keep in mind when it comes to any of these settlements, anything that Freddie or Fannie would do with respect to how they treat delinquent loans the GSEs post-2010 have a policy of pulling out loans at 120 days of delinquency.

So any of the loans that will be -- even if they were working them out differently, really won't affect prepayments on the underlying securities as they are already being pulled out.

Great, thank you very much.

Douglas Harter, Credit Suisse.

I was just hoping you could talk a little bit about your hedging strategy now that the Fed said that they are going to be on hold for longer, if anything changes with that.

Sure.

Look, I think what Peter outlined is a direction that we very likely will continue to go which is we like buying.

We feel that obviously four or five years in interest rates are going to be reasonably well pegged, so let's face it, what is the risk factor on interest rates.

It's really beyond the five year that the curve could steepen and inflation fears or other factors, such as dollar-related issues, could push the backend of the curve up.

And so we are kind of -- we like the hedging strategy of kind of buying out-of-the-money put options as being the key thing to do to protect book value against that kind of move.

So when you look at our position as it's evolving the prepayment protection that Chris described in the portfolio allows us to perform if rates stay low or go lower or in a QE3 environment.

The combination of the hedges that Peter outlined and specifically with the out-of-the-money put swaptions would certainly help in a rising rate environment.

So I mean just going forward would we expect to see more of the regular way swaps run off and be replaced by swaptions?

Yes, and you will see new swabs.

I don't want to tell you we are done using swaps.

It's clearly not the case; they will remain a major part of our hedge book.

But I think, to Peter's earlier point, it's very much a function of volatility.

We like the idea of using out-of-the-money options but we weren't going to buy any at the end of September and early October when volatility was through the roof, so to speak, and the price of those options was very expensive.

Option prices have come down a lot this year, and if option prices are pretty cheap, I think you can expect to see more of those.

Great.

Thanks, Gary.

Bose George, KBW.

Good morning.

I just wanted to revisit the issue of prepayments.

You noted your portfolio speeds are running at 8%, the life year assumptions are 14.

In terms of reconciling the two, should we assume that your view is that prepays are going to pick up pretty meaningfully in your portfolio?

Secondly, if it doesn't in a couple of quarters, do you have to kind of have a true down and bring it back down and reduce the amortization for one quarter?

So very good question.

The reality is the way we produce our projected prepayment speeds is -- and I have talked about this on prior calls.

It's not like the management team gets in a room and decides 14 looks like the right number.

We use a very well-known BlackRock Solutions third-party solution to come up -- or system to come up with our prepayment estimates.

They are driven by factors such as obviously interest rates, mortgage prices, house prices, all of the factors that we all talk about.

And they make updates to their models periodically to keep them up to date, so there are things like that that could happen.

But one thing that you want to keep in mind with respect to the prepayment estimates is that we do have some newer mortgages and newer mortgages as they season tend to pick up in speed.

So a brand-new mortgage that is one or two months old typically pays at very low CPRs.

Well, we don't like the idea of assuming a 2 CPR because it paid there in the second month the life of the security.

We want to use a longer run number and that is one of the things that is captured in those differences.

Now there are other things captured in those differences, such as how HARP loans prepay versus kind of reality, and those are the kinds of things that we will revisit over time as we get new information and as the market environment changes.

Okay, great, thanks.

That makes a lot of sense.

And then just actually had a macro question.

There is an article in The Journal today just on potential changes to the money market industry.

Curious if you had any thoughts about how any changes to that industry could potentially impact the repo funding markets, if you think there is any crossover there.

We are really not concerned about it.

Again, the repo market is extremely deep.

The reality is that there is plenty of short-term money on the sidelines right now, and agency repo or US government type securities are a necessary thing for people to invest in, both money funds and banks and plenty of other institutions.

We have just seen kind of really good stability there.

So the short answer is, no, we are not concerned.

Okay, great.

Thanks a lot.

Joel Houck, Wells Fargo.

Gary, just a follow-up on the higher CPR assumption.

Is another way of looking at that is that the average age of loans in your portfolio is newer than it was several quarters ago?

Because if you look at -- that seems to be the only logical explanation because the prepayment speeds for Fannie and Freddie now have been coming down a couple months.

Obviously your own portfolio is well-behaved.

So it seems like it's either really conservative or some other shift in the portfolio that we don't see in the slide presentation.

Look, I mean the portfolio is newer so that is a factor, but I do want to also point out that while prepayment speeds actuals have slowed down over the last couple months, as Chris said in his prepared remarks, mortgage rates are now back down at their all-time lows again.

And so it may be we are not going to look at the actual speed we just saw two months ago -- based on rates two months ago when mortgage rates were maybe 40 basis points higher.

So practically speaking there is a combination of the fact that there are newer mortgages.

Another piece is that we have some unique types of mortgages that continue to perform very well and may be performing well relative to model estimates.

But also keep in mind that mortgage rates have come back down.

Right.

And I guess maybe to follow-up on it, Gary.

Is it possible that the BlackRock model, because it's obviously based on historical relationship between changes in rates and refis or prepays, is perhaps overly conservative because of all the friction costs?

I hear what you are saying about rates are now at historic lows, but we have been in a very favorable prepayment environment now for quite a while and we are just not seeing the follow-through.

Perhaps that changes, but I guess the question is is there any consideration that the model is too conservative?

Look, we are going to continuously review the model along -- and BlackRock reviews it, our auditors look at it, and there is clearly tests for reasonableness.

And we will continue to monitor it as we get new data.

Let me give you an example where focus over time, over the next quarter or two will be higher, which is the best example are HARP securities.

The prepayment advantages afforded by HARP securities are not something that is easy to model looking backwards.

It's a new program.

A loan that goes -- the securities we are buying have gone through the HARP program and they don't have another refinance option.

So while BlackRock or we or anyone can make a projection about how that is going to perform as interest rates change, there is no reliable history to look at.

So as we continue to get informed on prepayment speeds on a newer product like that then that is an area where you are more likely to see model adjustments.

Lower loan balance, as an example, is something that has been there a long time and so some pretty good adjustments are being made.

Are they perfect?

No.

Could they be revised?

But I think you are likely to see the bigger revisions, so to speak, to things like the HARP securities.

And just to look quickly at the numbers, as an example, they are nine months old on 30-year.

We had $12 billion or $12.5 billion of HARP 30-year securities where the actual one-month average speed was 3 CPR and our projection is 11.

We are clearly going to have to look at that over time and make sure that as we get new information we are updating our estimates.

All right, great.

Thanks, Gary.

Martin DeVries, Barclays Capital.

Thanks.

Given the strength in the prepayment protected securities you laid out in the presentation, have you seen enough richening in those forms of payouts that now on a relative value perspective those types of investments on kind of new or reinvested money are less attractive?

This is Chris.

Valuations have improved but we still, generally speaking, like the risk-adjusted returns afforded by these strategies.

We did sell some of these pools during the fourth quarter versus both other call-protected pools and also lower coupon 20-year and 30-year pass-throughs.

But it's important to recognize that even within these specified pool categories there are certain attributes that can drive materially different performance.

For example, some servicers are significantly more efficient at reaching borrowers when there is an incentive to refinance.

In the case of HARP securities loans backed by or securities backed by loans that were originated through the HARP program, while there are different pricing relationships and different payups for different LTV buckets, two pools with the exact same LTV could have very different performance if the LTV, if the loan-level LTV distribution is very different.

And so we will continue to look for ways to upgrade and improve our positions, and I think the fourth quarter is a good example of that.

Okay, thanks.

I am sorry if you already addressed this, but were you surprised that HARP 2.0 seemed to have little to no effect on the February prepayment speeds?

Not really.

It shouldn't have had an impact in February.

If you really think about it, the changes were implemented we will call it late December.

Implemented meaning that people would start taking applications.

Some of the key changes aren't even in Freddie and Fannie's systems there would have been almost no time to close a HARP loan in January, even if they were in the systems, and some of the bigger changes, such as the use of automated valuations or appraisals versus actually having an appraiser out there haven't even been implemented now.

So I think it's just a fallacy I mean to have expected the impacts.

I mean that is why what we said and what we continue to believe is that you will start to see those effects really over the next couple of months, but I wouldn't expect to see a one-time pop.

I think what you will see is as systems get updated, both at Fannie and Freddie and on the part of the originators, that any impacts of HARP 2.0 will sort of gradually be phased in over the next three or so months.

Okay, thanks.

Steven Delaney, JMP Securities.

Good morning, everyone.

Just a couple of quick things.

First, on repo, on page 12 in your deck you are indicating that you believe that repo costs have come down close to 10 basis points from the 40 basis points at the end of the year.

We have heard of some quotes in the high 20s.

Could you give us a little color sort of -- don't know if you are just trying to be conservative -- close to 10; maybe what you are seeing as a range for 30- to 90-day quotes currently?

Hi, Steve.

It's Jason Campbell.

Right now we are averaging about 30 on one month, ranging probably the mid 20s to low 30s.

Okay, great.

And then three months is probably about 32 to 34.

Okay, great.

Because ICAP is showing a 25, 26 on Bloomberg, and I know that is not the real market but it's helpful to know there is some funding available inside of 30 basis points.

Yes, there are people at those levels but more generically it's around 30.

Understood.

Okay, thank you for that.

Then the final thing I had is this is a question I have been meaning to ask for some time and we can take it offline.

But is it possible for you to break out of your premium amortization of $121 million, is it possible to give us the amount that is associated with your long IO positions?

Why don't we take that offline, but the bottom line is our IO position is absolutely tiny at this point.

We had a larger long IO position and we were glad we had it back in the first quarter and into the beginning of the second quarter.

But as we told you, we have really reduced that position and at this point it's not really material.

Okay, so we should just assume -- I think it's around a couple hundred million cost from one of your slides -- but what you are saying is the vast majority of that is associated with the pass-throughs.

Yes, you are not looking -- that is just not going to be a driving factor.

All right, very good.

Thanks for that.

Matthew Howlett, Macquarie.

Good morning.

Thanks for taking my question.

Gary, just again on prepayment and I want to circle back to potential policy changes.

I know HARP 2.0 is in the books and I know that Obama would like to expand HAMP and do this FHA short refinancing program, which I guess would be mainly nonagencies.

But my question is do you think -- is your risk to policy change that, one, HARP 2.0 could be changed again which could, one, move the cut-off date from May 2009 to more [resubmitages] or, two, allow HARP 1.0 loans to refinance to new loans.

Is that the biggest policy risk?

And what do you see Obama I guess -- he hasn't really announced a lot of details.

What do you think is going to be said from him?

Well, I think we have gotten a feel for what he would like to do so to speak and there clearly is ambiguity there.

I think there is a general perception, both on our part and on the part of the mortgage market, that the material changes to GSE underwriting and GSE programs are over with.

The president gave his speech there was a day maybe where mortgage prices, higher coupon mortgage prices were impacted.

Since then they have done extremely well.

I think more instructive maybe than our opinion, which again is that we are passed most of the stuff, the market seems to believe that -- it's kind of tired of hearing about this and it views there is very little risk of anything material getting through Congress.

So I will start with that piece.

One thing that people don't realize is the Treasury and HUD, outside of just FHFA, were integral in the process to update to HARP 2.0.

This wasn't everyone else complaining about FHFA on HARP 2.0.

Treasury and HUD were at the table and you can see quotes and clear involvement on that front.

So keep that in mind.

Now just going to what is a worst-case scenario, yes, it probably is that the -- basically any loan is available for the HARP program.

Why is that a bad scenario?

Because then it means every loan is available for a streamline, a very, very streamlined refi program.

But remember we would have, in a sense, incremental risk that the HARP 1 securities maybe could be refinanced in that scenario so we understand that risk.

And that is exactly why we manage concentration risk.

But one thing on the other side of that is lower loan balance becomes even more important in that situation.

So if every loan is available for a streamlined refinance then what ends up becoming the key constraint?

Capacity in the system.

And what is a mortgage originator or a loan officer going to do when their only issue is how many loans they can process?

The first thing they are going to do is they are going to focus on larger loans because they get paid a percentage off of that loan.

So while the HARP securities may become more normal in that scenario, the low loan balance securities in some ways become even more valuable on a relative basis because they are a really bad kind of thing to focus on for the origination society so to speak.

So big picture, look, it's not something we would like to see.

We think it's a very low probability, but I want to stress that the composition of our portfolio, both in terms of coupons, in terms of loan balance, is diversified and set up because there are some risks that are outside of our control.

And just to be clear, they have not allowed loans originated after May 2009 to refinance through HARP, and they have not allowed HARP 1.0 loans to refinance?

That is still off the table right, as of now?

Absolutely.

No one is looking that; except if some bills get past could that change.

Yes, but absolutely it is not allowed at this point.

But what I want to make very clear is that that was given a lot of discussion and thought when the HARP 2.0 changes were made.

There were some very logical reasons why FHFA, HUD, and Treasury decided not to change that date.

As Chris mentioned to you and as we show you in the slides, there are substantial we say payups or premiums for the HARP securities.

Well, what do those premiums do?

Thos allow borrowers who go through the HARP program now, the people that the GSEs want to reach, they allow them to pay less points and less costs and have a lower rate on a new HARP loan.

So when you think about it that is a very important reason why it may not be in the best interest and probably isn't in the best interest of the GSEs or of even others in the government to allow HARP loans to re-HARP.

Because that payup will be much smaller.

So, again, something they gave a lot of thought to, something they left in place.

It would take a pretty big change at this point, probably be an act of Congress to change that.

Just on that -- a last follow-up on that question.

There has been suggestions that Obama could just replace DeMarco; he has been holding this process up apparently through a recess appointment.

What do you handicap that at?

I think what is really important is that -- I am not going to try to handicap DeMarco's job security.

I think what I would rather say is that I think there is a misconception out there that DeMarco as an individual is the sole reason that the HARP program isn't broader, so to speak.

And DeMarco is operating under essentially orders from Congress, which is that he is supposed to preserve the GSEs, and also he is supposed to be focused on the US mortgage market and its liquidity and its viability and so forth.

And again, HUD and Treasury were very involved in the HARP program.

I think they absolutely recognize a lot of these things that we are talking about.

Lastly, there is an Inspector General that Congress has appointed that oversees Fannie, Freddie, and FHFA who is very active and who is there to ensure that whoever is the head of FHFA is abiding by the laws and the orders that he is given, which is to protect the GSE's bottom line and not just to act as an arm of the administration.

So I think there is a misconception about in a sense how critical DeMarco is in that outcome.

Great.

Thanks, Gary.

Jim Young, West Family Investments.

Gary, you and your team have obviously navigated the markets very well over the course of the last year but you started the year with assets of about $13.5 billion and ended the year with about $55 billion in the portfolio.

So my question is at what size of your portfolio do you feel comfortable that you and your team can continue to navigate the changes in the markets, to continue to produce the attractive risk-adjusted returns that you have done in the past?

Thank you.

Look, thanks for the question.

What I would say is if you look at -- we are obviously a lot bigger than we were in 2009 and 2010.

One thing just on the surface is our economic returns, which is what we have told you to focus on for awhile, are actually stronger in 2011 than they were in 2010.

If you look at our performance versus our peers, if you look at our prepayments on our portfolio, if you look at the positioning relative to HARP 2.0, I think the facts are that we continue to be able to manage that.

I think to the other side of that question, which is is there a limit; the answer is, yes, there is a limit.

But remember that this is a $5 trillion market where there is, give or take, likely to be $1 trillion of originations every year.

So as big as agency could become, it's still going to be a tiny percentage of the market.

But the reality is there is a limit and we are not going to -- we evaluate our decisions around growth from the perspective that at this point size is not -- it can make sense if offerings are accretive, if there is good opportunities to buy good assets.

But you don't just do them for the heck of it, so to speak.

And I think if you look over the last month there was a lot of assumptions that AGNC was going to raise equity during the quarter or during January.

We chose not to and a big driver of that was we didn't feel that it was the right thing given a number of factors, including the growth in book value both last quarter and what we are seeing so far this quarter.

Thanks.

Can we go to the next question?

Mike Widner, Stifel Nicolaus.

Good morning, guys, or good afternoon I guess at this point.

Two quick ones; most of my questions have already been addressed here.

But just following up on one of the things you have mentioned in the course of conversation at the outset, Gary, was you guys do have a history of sort of prebuying when you find MBS opportunities that are attractive and then raising the capital.

You noted specifically, though, that last quarter the capital raise kind of mid-quarter was an exception to that.

Just wondering if you could talk a little bit about how you see the opportunities for prebuying.

We have had a little bit of pullback mid, late January in MBS prices on some recent strength, but all things considered yields are still attractive.

So how do you feel about the prebuying opportunities now and recently?

I think that is -- the reality is I will just go back to what I said earlier.

I will give you kind of two separate answers that are as close as I can get to your question.

First off, returns are reasonably attractive if you can find the right assets now, but they are not as attractive as they were three or four months ago.

And I think that is pretty straightforward.

The other thing is that what we have told you is that -- we have obviously disclosed a book value number.

We have also told you that mortgages have performed well this quarter and our specific holdings have performed well.

Those are obviously both factors that go into any decision on capital raising as well as other factors -- our view on the market, our views on Q3 and its likelihood, things like that.

Really outside of that there is not much more I can add.

Okay, got you.

You have got some new faces over there and I think one of the interesting things maybe for one of those guys, and you can choose who, is I think one of the finest pieces of explanatory footnotes I ever read is Note 5 on the bottom of page 24.

I thought maybe it would be interesting if you could ask one of the new guys to maybe explain that one for those of us that didn't find it completely transparent.

Which one was that again, can you repeat it?

It's footnote 5 on the bottom of page 24 that goes through the book value calculation.

What we are trying to describe there refers to the table.

If you think about the capital raise impact, it can have really two impacts.

It can have the price that we issue stock relative to the current book value.

That is the first line, that is the accretive nature of the capital raise which was $0.09.

That line that you refer to represents the kind of negative drag that the incremental shares had on our overall equity.

So think about those two net together with a $0.06 drag to our equity.

So I mean the line that -- the note that I am talking about is note 5 at the bottom of the page, which is the negative $0.15 reference there.

But that was a good piece of sort of explanation.

That is Pete, is that right?

Yes.

Well, Pete, I think I know exactly what the line item is, which is sort of the plug that is necessary to make all the other numbers jive, but the explanation for it I think is priceless there and the way you guys put that together.

Your proximity to Washington, DC, and legislation might have helped your wording on that one a bit.

We can take that one offline if you want to discuss it further.

The one thing I just want to -- on this subject, just to close out on, is again this basically applies accretion based on the closing book value.

As we said earlier, book value -- we track it in for a quarter and mortgage prices were a fair amount lower.

We are extremely confident that the equity raise was accretive and I think it's pretty obvious after the fact that it was good timing.

I don't doubt that at all.

It was really more specifically the language of that footnote that was wonderful.

Thanks, guys, and good solid quarter.

Daniel Furtado, Jefferies.

Dan, you there?

All right.

Well, thank you guys for your interest in AGNC and at this point we will close the call.

This concludes today's conference.

You may now disconnect.